D22496: Autmatic Test For Two Way ANOVA.

Stefan Gerlach noreply at phabricator.kde.org
Wed Jul 17 21:16:35 BST 2019


sgerlach added inline comments.

INLINE COMMENTS

> HypothesisTest.cpp:304
>  					(n[1]-1) * gsl_pow_2(std[1]) ) / df );
> -			statisticValue = (mean[0] - mean[1]) / (sp * qSqrt(1.0/n[0] + 1.0/n[1]));
> +            statisticValue.append((mean[0] - mean[1]) / (sp * qSqrt(1.0/n[0] + 1.0/n[1])));
>  			printLine(9, "<b>Assumption:</b> Equal Variance b/w both population means");

I would suggest to use sqrt() from cmath instead of qSqrt(). What do you think?

> HypothesisTest.cpp:747
>  
> +    double ss_a = 0;
>      for (int i = 0; i < np_a; i++)

you are doing a lot of basic statistical calculations. Can you check if any GSL function can be used (see https://www.gnu.org/software/gsl/doc/html/statistics.html)?

> HypothesisTest.cpp:843
> +    level++;
> +    rowMajor.append(new Cell("Interaction", level, true));
> +    rowMajor.append(new Cell(round(ss_interaction), level));

translation?

> HypothesisTest.cpp:849
> +    level++;
> +    rowMajor.append(new Cell("Within", level, true));
> +    rowMajor.append(new Cell(round(ss_within), level));

translation?

> HypothesisTest.cpp:1375
> +    if (pValue[0] > 1)
>  		return 1;
> +    return pValue[0];

check indentation

REPOSITORY
  R262 LabPlot

REVISION DETAIL
  https://phabricator.kde.org/D22496

To: devanshuagarwal, Stefan, asemke
Cc: sgerlach, kde-edu, narvaez, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20190717/f7a4892a/attachment.html>


More information about the kde-edu mailing list